Metraipolis › archive › dc_aacf2fdb2639
Resource-Efficient Reinforcement Learning Post-Training through Elastic Expert Parallelism — arXiv:2611.06984
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.06984
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T14:41:30.991Z
- SHA-256 of the archived text
a41fd918bb8662202776ebc5e1d904ba6feaed576a77da49efaef1d3894eb2a6- Stable URL of this page
https://metraipolis.com/documento/dc_aacf2fdb2639
How to cite
Resource-Efficient Reinforcement Learning Post-Training through Elastic Expert Parallelism — arXiv:2611.06984 — arXiv (en), https://arxiv.org/abs/2611.06984 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_aacf2fdb2639 (Metraipolis archive dc_aacf2fdb2639, sha256 a41fd918bb8662202776ebc5e1d904ba6feaed576a77da49efaef1d3894eb2a6)
BibTeX
@misc{dcaacf2fdb2639,
title = {Resource-Efficient Reinforcement Learning Post-Training through Elastic Expert Parallelism — arXiv:2611.06984},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.06984},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_aacf2fdb2639; sha256 a41fd918bb8662202776ebc5e1d904ba6feaed576a77da49efaef1d3894eb2a6},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06984. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reinforcement learning post-training improves the capabilities of large language models, but generating training responses is a major computational bottleneck. In mixture-of-experts models, workers that finish their assigned responses may remain occupied because their experts are still needed by workers processing longer responses. Consequently, approximately one-third of occupied worker time in the baseline occurs after local response generation finishes. Here we present FlexFloor, a system that enables earlier resource release during synchronous response generation while preserving the expert computation required by ongoing responses. FlexFloor jointly adapts response scheduling and the active group of workers. It uses historical response lengths and memory constraints to guide batching and prompt placement, then progressively releases workers while maintaining expert availability and correct communication. Compared with the verl baseline, FlexFloor generates approximately 65.40% more response tokens per unit of occupied computing time and increases response generation throughput by 52.74%, with little observed variation in rollout reward. By making reinforcement learning post-training more efficient, FlexFloor supports faster development and iterative improvement of large language models and AI agents. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Distributed, Parallel, and Cluster Computing (cs.DC). Pagina del lavoro: https://arxiv.org/abs/2611.06984 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-06T14:18:42.634Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).